Novel Genetic Loci Identified for the Pathophysiology of Childhood Obesity in the Hispanic Population Anthony G. Comuzzie1, Shelley A. Cole1, Sandra L. Laston1, V. Saroja Voruganti1, Karin Haack1, Richard A. Gibbs2, Nancy F. Butte3* 1 Department of Genetics, Texas Biomedical Research Institute, San Antonio, Texas, United States of America, 2 Human Genome Sequencing Center, Department of Molecular & Human Genetics, Baylor College of Medicine, Houston, Texas, United States of America, 3 USDA/ARS Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Houston, Texas, United States of America Abstract Genetic variants responsible for susceptibility to obesity and its comorbidities among Hispanic children have not been identified. The VIVA LA FAMILIA Study was designed to genetically map childhood obesity and associated biological processes in the Hispanic population. A genome-wide association study (GWAS) entailed genotyping 1.1 million single nucleotide polymorphisms (SNPs) using the Illumina Infinium technology in 815 children. Measured genotype analysis was performed between genetic markers and obesity-related traits i.e., anthropometry, body composition, growth, metabolites, hormones, inflammation, diet, energy expenditure, substrate utilization and physical activity. Identified genome-wide significant loci: 1) corroborated genes implicated in other studies (MTNR1B, ZNF259/APOA5, XPA/FOXE1 (TTF-2), DARC, CCR3, ABO); 2) localized novel genes in plausible biological pathways (PCSK2, ARHGAP11A, CHRNA3); and 3) revealed novel genes with unknown function in obesity pathogenesis (MATK, COL4A1). Salient findings include a nonsynonymous SNP (rs1056513) in INADL (p = 1.2E-07) for weight; an intronic variant in MTNR1B associated with fasting glucose (p = 3.7E-08); variants in the APOA5-ZNF259 region associated with triglycerides (p = 2.5-4.8E-08); an intronic variant in PCSK2 associated with total antioxidants (p = 7.6E-08); a block of 23 SNPs in XPA/FOXE1 (TTF-2) associated with serum TSH (p = 5.5E-08 to 1.0E- 09); a nonsynonymous SNP (p = 1.3E-21), an intronic SNP (p = 3.6E-13) in DARC identified for MCP-1; an intronic variant in ARHGAP11A associated with sleep duration (p = 5.0E-08); and, after adjusting for body weight, variants in MATK for total energy expenditure (p = 2.7E-08) and in CHRNA3 for sleeping energy expenditure (p = 6.0E-08). Unprecedented phenotyping and high-density SNP genotyping enabled localization of novel genetic loci associated with the pathophysiology of childhood obesity. Citation: Comuzzie AG, Cole SA, Laston SL, Voruganti VS, Haack K, et al. (2012) Novel Genetic Loci Identified for the Pathophysiology of Childhood Obesity in the Hispanic Population. PLoS ONE 7(12): e51954. doi:10.1371/journal.pone.0051954 Editor: Dana C. Crawford, Vanderbilt University, United States of America Received August 15, 2012; Accepted November 7, 2012; Published December 14, 2012 Copyright: ß 2012 Comuzzie et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by grants from the National Institutes of Health (NIH) (DK080457), and the USDA/ARS (Cooperative Agreement 6250-51000- 053). Work performed at the Texas Biomedical Research Institute in San Antonio, Texas was conducted in facilities constructed with support from the Research Facilities Improvement Program of the National Center for Research Resources, NIH (C06 RR013556, C06 RR017515). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected] Introduction chronic inflammation, and risk for fatty liver disease [3,4]. Identification of genes underlying these distinct patterns of Obesity is a complex disease influenced by genetic and association also may unravel important biological pathways environmental factors and their interactions. The current surge involved in the pathophysiology of childhood obesity. in childhood obesity in the U.S. is attributable to an interaction Genome-wide association studies (GWAS) have the potential to between a genetic predisposition toward efficient energy storage localize genetic loci contributing to obesity down to a few 100 kb and a permissive environment of readily available food and [5]. In fact, a recent meta-analysis of the adult GIANT sedentary behaviors [1]. Genetic architecture of common poly- Consortium established 32 susceptibility BMI loci [6], several of genic childhood obesity remains largely unknown. In genetic which were confirmed in French and German children with studies, the phenotypic description of the obese child usually has extreme obesity [7] and European adults with early-onset obesity been limited to body mass index (BMI). BMI represents [8]. Two novel loci near OLFM4 and within HOXB5 were recently a composite trait of fat free mass (FFM) and fat mass (FM) and reported based on a meta-analysis of 14 pediatric studies of BMI thus loci influencing BMI may differ from more direct measures of [9]. These pediatric GWAS were confined to BMI and cohorts of adiposity. In addition, markers of biological processes underlying European ancestry. the development of obesity such as dietary intake, energy Here, we present findings from a GWAS designed to identify expenditure and nutrient partitioning may be more effectual in genetic variants influencing childhood obesity and its comorbid- identifying causal genetic variants [2]. In epidemiology studies, ities in the Hispanic population. We have published evidence of childhood obesity has been shown to be genetically correlated with heritability, pleiotropy amongst traits, and chromosomal regions glucose intolerance, hypertension, dyslipidemia, insulin resistance, implicated in obesity among Hispanic children in our VIVA LA PLOS ONE | www.plosone.org 1 December 2012 | Volume 7 | Issue 12 | e51954 Obesity GWAS in Hispanic Children FAMILIA Study [4,10–15] In-depth phenotyping was performed procedure. The average call rate for all SNPs per individual to characterize the children, including anthropometry, body sample was 97%. composition, growth, metabolites, hormones, inflammation, diet, SNP genotypes were checked for Mendelian consistency using energy expenditure and substrate utilization and physical activity. the program SimWalk2 [24]. The estimates of the allele Our high-density SNP genotyping and phenotypes representing frequencies and their standard errors were obtained using SOLAR not only adiposity, but also biological processes associated with the [25]. Specific SNPs were removed from analysis if they had call development and consequences of childhood obesity enabled rates ,95% (n = 6,596), were uncommon alleles in less than 5 localization of novel genetic loci associated with the pathophys- participants (26,537), deviated from Hardy-Weinberg equilibrium iology of childhood obesity. (n = 0), or were monoallelic (n = 56,448). The number of SNPs that passed quality control and were included in the GWA analysis Materials and Methods was 899,892. The VIVA LA FAMILIA Study was designed to identify genetic Genome-wide Association Analysis variants influencing pediatric obesity and its comorbidities. Family Measured genotype analysis (MGA) was performed using the recruitment and phenotyping were conducted in 2000–2005 in SOLAR program [25]. All phenotypes were transformed by Houston, TX. All enrolled children and parents gave written inverse normalization to meet assumptions of normality. We informed consent or assent. The protocol was approved by the obtained residuals using linear regression models adjusted for age, Institutional Review Boards for Human Subject Research for sex, their interaction and higher order terms. Because energy Baylor College of Medicine and Affiliated Hospitals and for Texas expenditure is strongly influenced by body weight across the age Biomedical Research Institute. range of 4 to 19 years of age, total energy expenditure and sleeping The VIVA LA FAMILIA study design and methodology have energy expenditure were additionally adjusted for body weight. been described in detail elsewhere4. GWAS was performed on 815 Also, observed energy intakes consumed in a snack and dinner children from 263 Hispanic families. The number of families by were adjusted for total energy expenditure or estimated energy sibships was: 8 (one child), 40 (two children), 155 (three children), requirement, again to compensate for the wide range of ages in 48 (four children), 6 (five children), 3 (six children), 2 (seven our cohort. children) and 1 (eight children). Each family was ascertained on an Each SNP genotype was converted in SOLAR to a covariate obese proband, defined as a BMI.95th percentile, between the measure equal to 0, 1, or 2 copies of the minor allele (or, for ages 4–19 y. Once identified, the obese proband and all siblings, 4 missing genotypes, the weighted covariate based on imputation). to 19 y of age, and their parents were invited to the Children’s These covariates were included in the variance-components mixed Nutrition Research Center for a tour and full explanation of the models for measured genotype analyses [26] versus null models study prior to consenting. The cross-sectional, longitudinal study that incorporated the random effect